Abstract Prediction of seed vigor based on hyperspectral peant. The traditional method is time-consuming and laborious to detect seed vigor. At the same time, the accuracy of the detection result is not high, and it will cause damage to the seed itself. Therefore, in order to achieve rapid and non-destructive detection of peanut seed vigor, the test was performed with original health, artificial aging for 24h and Peanut seeds with different vigor gradients at 72 hours were used as the research samples. Hyperspectral images with a wavelength range of 387~1035 nm were collected, and the image of the central part of the peanut seeds with a pixel size of 60 × 60 after correction was intercepted and the average reflectance value was calculated. After a combination of processing analysis, characteristic band processing, and model selection, a hyperspectral prediction system with the highest correlation to the viability of extracted peanut seeds was finally established. Experiments shown that the combination of hyperspectral imaging technology and the MF-LightGBM-RF model had the best performance, with a prediction accuracy of 92.59% and a fitting time of 1.77s, which simplifies the model and improves efficiency.

Abstract This study investigated the hyperspectral reflectance response of time series generated during oven drying to changes in the moisture content of potato tubers. Seventeen preprocessing methods were used to eliminate the influence of spectral noise on the spectral characteristic curve. Algorithms such as CatBoost, LightGBM, and XGBoost are used to obtain the first 40 effective characteristic spectra of hyperspectral images, which reduces the redundancy of data and improves the prediction accuracy. The water content prediction model of potato tubers was established by using the selected characteristic bands. The results showed that the combined model based on Lasso and XGBoost algorithm had the strongest prediction ability. The best model is MF-Lasso-XGBoost, which has R2 value of 0.8908, Rmse of 0.0610, Mdae of 0.0389, and R2cv of 0.8448. This research can provide reference for the detection of potato moisture content and theoretical basis for the development of crop moisture detector.

Abstract Peanut storage time affected the quality of peanut seed sowing and germination and also affected the taste of edible peanuts. With the increase of peanut storage time, the total amount of water and amino acids decreased, and peanuts appeared moldy. The artificial judgment of peanut storage time mostly relied on visual classification to evaluate the color, which leads to large differences in color classifications between observers. This research was conducted to determine the fresh state of peanuts during storage based on the hyperspectral imaging (HSI) technology, and to identify the storage time of peanuts through hyperspectral images (387~1035 nm). Three models, two preprocessing methods, and two feature band extraction methods were combined. The experimental results shows that the DT-MF-Catboost model was the best method to detect the storage time of peanuts, and its accuracy of identifying the storage time of peanuts was 97.53%. Studies have shown that HSI has great potential in classifying the freshness and identification of peanuts, and provides a basis for non-destructive testing classification as well as grading of peanuts during storage.

Abstract In order to realize the rapid nondestructive detection of mildew peanut in the process of peanuts storage, hyperspectral imaging technology was proposed to detect mildew peanut. A total of 200 peanuts were selected from 5 kinds of peanuts purchased in the market for moldy treatment, and the remaining 400 peanuts were kept sterile. After completion, samples were collected with a hyperspectral instrument to obtain spectral data of the samples. According to the characteristics of the data, 10 pre-processing algorithms were used to de-noise the data, and Median Filtering (MF) had the best effect, with the recognition accuracy reaching 97.7%. GBDT, LightGBM, CatBoost and XGBoost algorithms were used to extract important feature bands in the spectral data pre-processed by MF. GBDT, LightGBM, CatBoost and XGBoost algorithms were used to model the extracted feature bands. The results showed that LightGBM is the best algorithm with a detection rate of 99.10%. Optuna algorithm was used to tune its parameters. Compared with the previous model, the running time of the optimized model was improved by about 0.25 s. The results showed that hyperspectral imaging provides an efficient and nondestructive method for detecting mildew in peanut storage.